Direct Three-Dimensional Visualization of Membrane Disruption by Amyloid ﬁbrils

Direct three-dimensional visualization of membrane disruption by amyloid ﬁbrils

Lilia Milanesia, Tania Sheynisb,c, Wei-Feng Xueb,1, Elena V. Orlovaa, Andrew L. Hellewellb, Raz Jelinekc, Eric W. Hewittb, Sheena E. Radfordb, and Helen R. Saibila,2

aDepartment of Crystallography and Institute of Structural and Molecular Biology, Birkbeck College, London WC1E 7HX, United Kingdom; bAstbury Centre for Structural Molecular Biology and School of Molecular and Cellular Biology, University of Leeds, Leeds LS2 9JT, United Kingdom; and cDepartment of Chemistry, Ben-Gurion University of the Negev, Beer-Sheva 84105, Israel

Edited by Jonathan S. Weissman, University of California, San Francisco, CA, and approved October 26, 2012 (received for review April 15, 2012) Protein misfolding and aggregation cause serious degenerative the number of fibril ends at a given protein concentration. Be- conditions such as Alzheimer’s, Parkinson, and prion diseases. Dam- cause these are known to be reactive sites, the above observations age to membranes is thought to be one of the mechanisms under- suggest a role for fibril ends in amyloid–lipid interaction and lying cellular toxicity of a range of amyloid assemblies. Previous possibly in amyloid pathogenesis (23, 24, 27). Fragmented fibrils studies have indicated that amyloid fibrils can cause membrane of all three proteins were also found to induce dye leakage from leakage and elicit cellular damage, and these effects are enhanced negatively charged liposomes, the most susceptible of which by fragmentation of the fibrils. Here we report direct 3D visualiza- contain a mixture of the cellular lipids phosphatidylcholine (PC) tion of membrane damage by specific interactions of a lipid bilayer and phosphatidylglycerol (PG), but liposomes with a variety of fi β with amyloid-like brils formed in vitro from 2-microglobulin compositions were damaged by the fibrils in all cases (27). β ( 2m). Using cryoelectron tomography, we demonstrate that frag- Here, we use β2mamyloidfibrils formed in vitro as a model β fi mented 2m amyloid brils interact strongly with liposomes and system to investigate the structural basis of membrane damage by cause distortions to the membranes. The normally spherical lipo- amyloid fibrils (27, 28). Previous studies have shown that these somes form pointed teardrop-like shapes with the fibril ends seen fibrils possess a parallel in register cross-β structure (29, 30) as- in proximity to the pointed regions on the membranes. Moreover, sembled into multidomain filaments coiled together, described by the tomograms indicated that the fibrils extract lipid from the mem- cryo-EM (28). These fibrils bind amyloid-specificligandssuchas BIOPHYSICS AND

branes at these points of distortion by removal or blebbing of the serum amyloid P component with a similar affinity to their ex vivo COMPUTATIONAL BIOLOGY fl – outer membrane lea et. Tiny (15 25 nm) vesicles, presumably formed counterparts (31). Using the conditions under which β2mamyloid from the extracted lipids, were observed to be decorating the fibrils. fibrils induce dye release from liposomes (pH 7.4), we examined The findings highlight a potential role of fibrils, and particularly fibril the effects of both long (∼1,400 nm) and fragmented (∼400 nm) ends, in amyloid pathology, and report a previously undescribed class β2m fibrils (27), as well as various control preparations, on the 3D of lipid–protein interactions in membrane remodelling. structures of the liposomes by confocal microscopy, cryo-EM, and tomography. We found pronounced distortions in the liposomes, he failure of molecular chaperones to prevent the accumu- interruptions to the bilayer structure, and extraction of lipids that Tlation of misfolded proteins results in protein aggregation were induced by the presence of amyloid fibrils. The most severe and amyloid formation, processes associated with severe human distortions were seen in proximity to the fibril ends, which are degenerative diseases (1, 2). Despite the attention focused on enriched in the fragmented fibril samples. This type of membrane these problems during the century since these disorders were first remodelling appears distinct from the actions of other previously identified (3–5) and advances in understanding the structure of described proteins that induce membrane breakage, as in the the cross-β conformation of amyloid fibrils in atomic detail (6, 7), action of membrane pore-forming proteins (32). The results the basic pathological mechanisms of amyloidosis remain poorly suggest a role of fibrils in membrane damage that could contribute understood and therapeutic intervention is lacking. The identity to the cellular dysfunction associated with amyloid disease. of the toxic species and the mechanisms of cytotoxicity remain Results major unsolved problems. In some systems, there is evidence suggesting that prefibrillar oligomers, rather than the fully formed Large unilamellar vesicles (LUVs) formed from 80% egg PC and fi fibrils, are the source of toxicity (8, 9). In these cases, cytotoxicity 20% egg PG, prepared by extrusion with a 100 nm lter, showed – is thought to result from the formation of specific membrane smoothly rounded shapes in cryo-EM images (size range 60 130 A Inset fi pores (10, 11) although alternative models including membrane nm; Fig. 1 , ). The wide eld view of the liposomes on a lacy fi destabilization or membrane thinning have also been proposed carbon support lm shows that the liposomes are sparsely dis- (12–15). In other cases, toxicity may reside with the amyloid fibrils tributed over the EM grid, mainly adhering to the carbon at the A β fi themselves. Evidence that toxicity correlates with fibrillar as- edges of holes (Fig. 1 ). When 2m amyloid brils initially semblies has been reported for yeast and mammalian prion formed at pH 2.0 are transferred to pH 7.4, they associate into proteins (16, 17), human lysozyme (18), Huntingtin exon 1, α- synuclein (19), and Amyloid-β (Aβ) (20, 21). Furthermore, Aβ plaques have been shown to form rapidly in vivo and to precede Author contributions: R.J., S.E.R., and H.R.S. designed research; L.M., T.S., W.-F.X., and A.L.H. performed research; R.J., E.W.H., and S.E.R. contributed new reagents/analytic neuropathological changes in a mouse model (22). The end sur- tools; L.M., T.S., W.-F.X., E.V.O., R.J., E.W.H., S.E.R., and H.R.S. analyzed data; and L.M., faces of fibrils (herein termed “fibril ends”) are unusually reactive T.S., W.-F.X., E.V.O., A.L.H., R.J., E.W.H., S.E.R., and H.R.S. wrote the paper. entities: they play a key role in catalyzing recruitment and con- The authors declare no conflict of interest. formational conversion of amyloid-forming proteins (23, 24) and This article is a PNAS Direct Submission. fi provide the sites for templated elongation of amyloid bril growth Freely available online through the PNAS open access option. fi β (25, 26). Recently, Xue et al. (27) showed that short brils of 2- 1Present address: School of Biosciences, University of Kent, Canterbury, Kent, CT2 7NJ, microglobulin (β2m), α-synuclein, and hen lysozyme, each pre- United Kingdom. pared by fragmentation of longer fibrils, cause increased damage 2To whom correspondence should be addressed. E-mail: [email protected]. to membranes and disruption to cellular function compared with This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. the initial long fibrils. Short and long fibril preparations differ in 1073/pnas.1206325109/-/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.1206325109 PNAS Early Edition | 1of6 Downloaded by guest on September 29, 2021 the criteria for classifying liposomes as pointed, smoothly rounded, A B or ambiguous are shown in Fig. S1. The ambiguous category was needed because half of the liposomes were partially obscured in the samples with short fibrils by contacts with fibril bundles and/or other liposomes (and one-third in the samples with long fibrils), preventing their reliable classification as either pointed or smooth. To determine whether the observations made are specificto the amyloid-like structure of the β2m fibrils, we examined the effect of a different type of β2m aggregate, known as “worm-like” (WL) fibrils, on the liposome preparations. These WL fibrils are formed under different conditions to the amyloid-like fibrils of β2m and do not contain a highly ordered cross-β structure, as revealed by EPR, magic angle spinning NMR, and FTIR (29, 30, C D 33). Moreover, they exhibit a reduced ability to disrupt cellular function and to induce dye release from liposomes compared with their amyloid-like counterparts (27). The WL fibrils also interact differently with egg PC/PG liposomes, showing only partial clustering of the liposomes as visualized by cryo-EM, with many liposomes remaining in regions free of fibrils (Fig. S2A). This suggests much weaker interaction of the WL fibrils with lipid compared to both the short and long amyloid-like fibrils, consistent with the liposome dye-release experiments described above. β2m monomers had no detectable effect on the distribu- tion or appearance of the liposomes (Fig. S2B), nor did control filaments that lack a cross-β structure such as microtubules (Fig. E F S2C) or tobacco mosaic virus (TMV) (Fig. S2D). To examine the effect of fibrils on membrane integrity in solution, fluorescence microscopy was used to image giant vesicles (GVs) labeled with green fluorescent lipid, mixed with short β2m fibrils labeled with 5-(and 6-)carboxytetramethylrhodamine (TMR) (Fig. 2). Fluorescence and bright field images of these liposomes incubated with buffer alone showed stable, round structures that remained intact with no visible deformation even after 3 h of incubation (Fig. 2A and Fig. S3A). Similar results were obtained when the vesicles were incubated with TMR-labeled β2m monomers (Fig. S3B). Images of short or long β2m β fi fi Fig. 1. Cryo-EM overviews of liposomes and 2m brils. Low-magni cation fibrils alone at pH 7.4 showed irregular aggregates, consistent β fi fi images of (A) liposomes, (B) short 2m brils, (C) liposomes plus short brils, with the bundling of fibrils observed by EM (Fig. S3 C and D), and (D) liposomes plus long fibrils. (Scale bar for A–D main images: 1 μm.) (E whereas small aggregates were observed for WL fibrils alone and F) Higher magnification views of liposomes with short (E) or long (F) E fi fi (Fig. S3 ). Mixing the liposomes and short brils resulted in brils. Examples of distorted liposomes are indicated by white arrowheads in severe damage to the liposomes (Fig. 2 and Fig. S4A). Strikingly, E. The inset in A shows a higher magnification view of liposomes. (Scale bar fi for E and F: 200 nm.) the surfaces of the bril aggregates colocalize with lipid (yellow regions), suggesting that lipid extraction from the liposome membranes had occurred upon mixing with the fibril sample. irregular bundles and aggregates that are distributed over the Less extensive damage occurred with long fibrils (Fig. S4B), entire grid (Fig. 1B), consistent with our previous observations whereas in the presence of WL fibrils many liposomes remained C (28). Mixing liposomes with either short or long amyloid-like intact (Fig. S4 ). These observations are consistent with the fi fibrils formed from β m results in densely clustered liposomes relative potencies of the different bril types in perturbing li- 2 A entangled with the fibrils (Fig. 1 C and D). In contrast with the posome structure as indicated by cryo-EM (Fig. 1 and Fig. S2 ), fi in dye-release experiments, and in causing cellular damage (short nding that the liposomes alone adhere poorly to EM grids and fi > fi >> fi appear mainly at the carbon edges, the fibril–liposome mixtures brils long brils WL brils (27). However, these approaches are densely distributed across the entire grid, suggesting strong cannot reveal the mechanism of membrane disruption, nor can they show the details of the fibril–liposome interaction. fibril–liposome interactions. Analysis of the liposome structure Although the cryo-EM images shown in Fig. 1 indicate in the presence of short, fragmented fibrils revealed that some a strong interaction of the amyloid fibrils with liposomes, it is not liposomes are distorted into extended, teardrop-like shapes (Fig. E possible to determine from individual projection images whether 1 , open arrowheads). In these samples, liposomes with altered apparent overlaps between fibrils and liposomes indeed indicate shapes are readily observed, and the liposomes are permeabilized a contact in the same plane of the specimen. Therefore, we used in dye release experiments under these conditions (27). Although cryo-electron tomography to examine the 3D structures at the fi the long brils also bind and concentrate the liposomes, in these sites of liposome interaction with short fibrils by reconstruction samples many liposomes remain smoothly rounded, even when from a series of images recorded at different tilt angles. The they contact large bundles of fibrils, consistent with the reduced results revealed examples of liposomes distorted into pointed, ability of the long fibrils to cause dye leakage (27) (Fig. 1F). To tear-drop-like shapes in every tomogram examined. These are quantitate the difference in frequency of liposomes with point shown in sections through representative 3D reconstructions distortions in samples incubated with short or long fibrils, we (Fig. 3 A–C). The fibrils typically form lateral contacts to the counted >5,000 liposomes. This revealed five times more point lipid surfaces and examples were seen in which the fibrils termi- distortions in samples incubated with short fibrils compared with nate in close proximity to sharp discontinuities in the liposomes. their longer counterparts (Table S1). Example images showing The tomogram region in Fig. 3C was used to trace the structure

2of6 | www.pnas.org/cgi/doi/10.1073/pnas.1206325109 Milanesi et al. Downloaded by guest on September 29, 2021 NBD-PE TMR- 2m Bright Field Merge A

20 µm

Fig. 2. Fluorescence microscopy showing liposome damage by short β2m fibrils. (A) Liposomes plus buffer and (B) liposomes plus short fibrils. Green, NBD-PE labeled giant vesicles; red, TMR-β2m fibrils. Yellow regions indicate colocalization of lipids and fibrils. In B, the vesicles are either completely disintegrated by fibrillar aggregates or show membrane damage with visible lipid extraction by the fibrils.

of the 100 nm liposome with associated fibrils and small vesicles, tomograms collected from three different samples. The inter- and these are shown as rendered surfaces superposed on a to- actions typically involve a stretch of lateral contact between the mogram section in Fig. 3D. Tiny vesicles, with diameters ranging liposome and the side of a fibril, terminating with a distortion or from 15 to 25 nm, were observed in contact with the clusters sharp discontinuity in the liposome shape at the end of the fibril. of short fibrils in 80% of tomograms examined (Fig. 3D, red Of the 244 interactions, 172 interactions (70%) involved contacts BIOPHYSICS AND

spheres). Although some vesicles in this size range are also of the vesicles with both sides and ends of the fibrils, 20 contacts COMPUTATIONAL BIOLOGY observed inside larger ones, presumably formed during lipid (8%) involved only the sides of the fibrils, and 55 contacts (22%) extrusion, free vesicles of this size range were never observed in were with fibril ends. An example of each of these types of in- samples that had not been incubated with amyloid fibrils. teraction is shown in Fig. 4 A–C. In general, the membranes were To quantify the nature of the observed fibril–vesicle inter- flattened at sites of contact with other vesicles, edges of the actions, we divided the contacts into three categories: fibrils carbon film and fibril sides, but showed sharper distortions in the binding the membrane surface by their sides, at their ends, and vicinity of the fibril ends. by a combination of both interactions. We counted the number The EM images suggest that the short fibrils bundle together E F of interactions in each category in a dataset of 244 clear exam- more than their longer counterparts (Fig. 1 and ). Bundling fi – ples (in 3D) of fibril–membrane interactions from four different of brils decreases the lateral surface volume ratio, and therefore the area of lateral surface available for interaction with liposomes. Nevertheless, the short fibrils, which have a higher proportion of exposed ends, increase dye permeability of liposomes (27), increase membrane disruption (Fig. 2), and cause A B more point distortions in the lipid bilayer in the vicinity of fibril ends (Figs. 3 and 4 and Movies S1 and S2). Together, these data suggest that the ends of the short fibrils are the most reactive species in membrane disruption, although the sides of the fibrils also play a role in membrane binding. Should fibril bundling be an important factor in the disruption of liposomes, the liposomes would be expected to form contacts to extended regions on the bundle surfaces—for example, the membrane might be expected to extend over, or wrap around, the bundle surface. However, such behavior was never observed in the tomograms. Instead, in C D all of the examples in which the liposome–fibril contact is clearly resolved (Figs. 3 and 4 and Movies S1 and S2), the membrane– fibril contacts are localized, so that the membrane appears to touch only one fibril in any given contact, even when that fibril is part of a bundle. These observations argue against a major effect of fibril bundling in the interaction with liposomes. To better understand the details of membrane damage by the short β2m amyloid fibrils, we collected examples of these interactions in tomograms recorded with a lower defocus, so that the lipid bilayer was resolved, albeit at the expense of lower contrast. 50 nm The examples shown in Fig. 4 D–F reveal the structural basis of how amyloid assemblies can extract lipid from membranes. –fi – Breaks and distortions appear in the outer leaflet of the bilayer, Fig. 3. Cryoelectron tomography of liposome bril interactions. (A C) fi fi Sections of tomograms showing liposomes clustered and distorted by short in regions adjacent to bril ends, suggesting a speci c interaction fibrils. (D) A rendered 3D model of a distorted liposome, surrounding fibrils that leads to destabilization and eventually disintegration of and adjacent small vesicles from C. (Scale bar: 50 nm.) the membrane.

Milanesi et al. PNAS Early Edition | 3of6 Downloaded by guest on September 29, 2021 ABC methods, Reynolds et al. (12) observed fibril growth and lipid extraction/thinning by adding monomeric α-synuclein to membranes supported on a solid surface. In a similar study it was shown that addition of islet-associated polypeptide (IAPP) induces defects on supported lipid bilayers, accompanied by transfer of lipid vesicles onto growing amyloid fibrils (14). These studies provide a large body of evidence suggesting a significant involvement of lipid membranes in amyloid cytotoxicity. Our results show that amyloid fibrils bind strongly to membrane surfaces, as shown by the marked clustering of liposomes in the presence of β2m fibrils. For many amyloidogenic proteins and amyloid fibrils, it has been shown that binding to membranes requires negatively charged lipids and positively charged residues or areas of high positive charge density on the protein assembly (12, 37–39). This is also the case for the interaction of β2m fibrils with liposomes: higher dye leakage was found for liposomes containing negatively charged lipids (27), suggesting that the fibril–membrane interaction has a strong electrostatic component. DE F In the case of IAPP, binding of the monomeric protein to membranes is mainly driven by electrostatic interactions, but the membrane damage is dominated by hydrophobic interactions with IAPP oligomers, similar to the membrane-mediated toxicity of other amyloidogenic assemblies (11, 21, 39). A key finding from the cryo-electron tomography images portrayed here is the sharp distortion of the membranes by the fibrils, suggesting that the fibrils make additional interactions that extract lipids from the outer membrane leaflet of the liposomes. Notably, initial studies of liposomes incubated with fragmented α-synuclein fibrils also showed examples of sharp distortions and interruptions to the bilayer at sites of contact with α-synuclein fibrils. The notion of lipid extraction by amyloid fibril ends is supported by the observation of tiny (15–25 nm) vesicles found near the areas of contact between β2m fibrils and distorted liposomes, and it is consistent with the colocalization of fibrils and disintegrated liposomes observed by fluorescence microscopy (Fig. 2B and Fig. S4 A and B). The observation that the sites of membrane disruption are mainly adjacent to fibril ends accords with the finding of the greatest liposome disruption by fibrils that were fragmented and therefore contained more Fig. 4. Distortions of liposomes in the vicinity of fibril ends. (A–C) Examples fi – ends, suggesting that the fibril ends have an enhanced ability and cartoons of the different types of bril liposome interactions observed, fi with the percentage of each type measured by counting examples in four (relative to the bril shaft) to cause the sharp distortions and tomograms. (D–F) Sections of tomograms taken closer to focus, showing extract lipids from the membrane. It is likely that these distor- examples of the disruption of the lipid bilayer in the region of fibril ends, tions involve hydrophobic interactions in addition to the elec- including (D) formation of a sharp point, (E) a break in the outer leaflet of trostatic component. Hydrophobic domains in proteins can the membrane, and (F) a bubble forming in the outer leaflet, with corre- induce dramatic distortions in lipid membranes. For example, sponding cartoons showing membranes as black lines and fibrils as gray the ability of epsin to induce membrane tubulation and vesicu- lines. (Scale bar: 50 nm.) lation is attributed to distortion caused by the insertion of wedge- shaped hydrophobic domains into the membrane (40). The fi Discussion hydrophobicity of the bril ends may be a property in common with prefibrillar or fibrillar oligomers, which are also reported to There is considerable evidence that amyloid assemblies interact damage membranes and are considered as intermediates in with membranes, supporting the notion that these interactions amyloid fibril assembly and disassembly (11, 41–43). may play a role in amyloid-associated cellular dysfunction (11, Membrane distortion and breakage occur in many biological 13, 34). However, the physical basis for these interactions and processes, such as viral fusion and pore formation, processes in the mechanisms of amyloid-induced dysfunction remain un- which binding of extraneous proteins induces membrane curva- resolved. Observations of ring-like structures formed from ture and breakage, resulting either in fusion with another mem- α-synuclein and prefibrillar assemblies of other amyloid-like brane or in the formation of protein-lined transmembrane systems, along with membrane leakage in the presence of prefi- channels (32, 44). The fibril–liposome interaction appears to re- brillar oligomers, have led to the hypothesis of membrane pore present a new class of specific membrane distortion by a protein formation, membrane thinning, and membrane destabilization assembly, distinct from previously described mechanisms for by prefibrillar oligomers (8, 9, 12, 34, 35). In addition it has been distorting and disrupting lipid bilayers. Similar distortions have shown that lipid-induced depolymerization of nontoxic Aβ fibrils been seen by cryoelectron tomography of liposomes involved in leads to formation of so-called “reverse oligomers” that are cy- a different process, the formation of viral fusion pores (45). In totoxic, akin to the oligomers formed de novo during fibril as- this case, formation of a pore and fusion of the membranes are sembly (36). Membrane-induced fibril depolymerization and directed by hemagglutinin spikes on the viral surface that bind membrane destabilization associated with fibril growth on lipid and insert into the target liposome membrane. Binding occurs bilayers have also been proposed as alternative mechanisms of via the ends of the spikes, which undergo a low pH-induced amyloid-mediated cytotoxicity (11, 13, 15). Using fluorescence conformational change to insert. In this case, a sharp point is

4of6 | www.pnas.org/cgi/doi/10.1073/pnas.1206325109 Milanesi et al. Downloaded by guest on September 29, 2021 also drawn out from the target membrane, as is observed with immediately before grid preparation. For each sample of liposomes and fibrils, grids were prepared in batches of three, within ∼7 min after addition short β2m amyloid fibrils, but instead of destroying the liposome, the destabilized, extruded membrane region is fused to the of fibrils to the liposomes. Liposome damage is complete within 10 min viral membrane. according to the dye release measurements (27). – The cryoelectron tomography images of β m fibril–membrane Control samples containing only the PC PG stock liposomes and samples 2 – β fi interactions presented here suggest that the cellular dysfunction of PC PG liposomes with added TMV, microtubules, WL 2m brillar aggregates, long unfragmented β m fibrils, or β m monomer were prepared associated with these and other fibrils or fibril-like assemblies 2 2 using the same protocol. Equivalent samples containing only fragmented, (16–21, 27) involves direct bilayer disruption. This membrane β fi fi fi WL, or long (unfragmented) 2m brils were prepared in parallel. The nal damage might arise by direct interaction with the bril ends and/ concentrations of TMV and microtubules in the samples containing PC–PG or by the creation of new, toxic species by reaction of the fibril liposomes were 3.5 mg/mL and 7 μM (tubulin monomer equivalent), re- fi ends with the lipids. The ends of biological laments differ spectively. For each control sample, the effects of fragmented β2m fibrils dramatically in their structure and properties, with actin fila- were examined on the same stock of PC–PG liposomes to control for the ment and microtubules being classic examples (46, 47). For inherent variability of extruded liposomes. Control samples of WL fibrillar amyloid fibrils, fibril ends are also distinct, being the sites of aggregates and PC–PG liposomes prepared according to the above protocol growth and disassembly, consistent with the notion that amyloid contained a higher salt concentration (196 mM NaCl), as required for fibril ends have a specific role in membrane disruption. The preparation of WL fibrillar aggregates (48). Therefore, samples of frag- β fi results presented here suggest previously undescribed routes of mented 2m brils and liposomes, as well as liposomes alone, were analyzed to verify that the higher salt did not affect the results obtained. Similarly, amyloid-associated cellular dysfunction involving membrane fi β disruption by fibril ends. These observations suggest, in turn, the pH 2.0 buffer used to assemble the amyloid-like brils of 2m had no effect on the liposome structure or integrity. that inhibiting fibril–lipid interactions by capping fibril ends may − provide a potential unique therapeutic strategy targeting amy- Cryo-EM, Tomography, and Image Processing. Low-dose (15–20 e /A2) images loid pathogenesis in disease. were recorded on a Gatan 4K × 4K charge coupled device (CCD) camera (Gatan) at a calibrated magnification of 29,000× (3.89 Å/pixel) and 2–3.5 μm Methods underfocus with a Tecnai F20 FEG electron microscope operated at 200 kV β fi 2m Fibril Preparation. Long straight brils were prepared from recombinant and equipped with a Gatan cold stage. Overview images were recorded at β 2m in 10 mM NaH2PO4, 50 mM NaCl buffer pH 2.0 containing 0.1% (wt/wt) 5,000×. Cryo-EM images collected from five different preparations of PC–PG fi fi bril seeds, and WL brillar aggregates were prepared in 25 mM NaH2PO4, liposomes showed that the liposomes were unilamellar, and more than 70% 400 mM NaCl buffer pH 2.5 following previously described protocols (27, 48). had diameters in the range 60–120 nm. No liposomes were observed with BIOPHYSICS AND fi Short fragmented bril samples were generated by stirring preformed long diameter less than ∼20 nm (Fig. S2). fi COMPUTATIONAL BIOLOGY straight bril samples at 1,000 rpm for 48 h at 25 °C using a precision stirrer Cryoelectron tomography was done on a Tecnai Polara electron micro- (custom-built by the workshop of the School of Physics and Astronomy, scope (FEI) operated at 300 keV with a calibrated magnification of 23,000× University of Leeds), as previously described (27). The protein monomer (5.1 Å/pixel). Tilt series covering an angular range of –60° to +60° at 2° concentration was 120 μM for the straight fibrils and 45 μM for the WL − increments and nominal underfocus of 2.5–5 μm and 70–90 e /A2 per series aggregates. Fibril samples were stored at 25 °C and used within 3 mo. To were recorded on a Gatan Ultrascan 4000 4K × 4K CCD camera using FEI prepare fibrils for confocal imaging, β m monomers were labeled with TMR, 2 tomography software. A total of 33 tilt series were collected from four as described in Porter et al. (49) and as detailed in SI Methods. different samples of PG–PC liposomes and fragmented β2m fibrils. The images were binned to 10 Å/pixel and aligned with gold bead fiducial TMV and Microtubule Controls. TMV (16 mg/mL) was supplied by J. W. M. van markers. Tomogram reconstructions were calculated by weighted back- Lent (Laboratory of Virology, Wageningen, Netherlands). Microtubules projection in IMOD (52). Some tomograms were also reconstructed from prepared by polymerization of glycerol-free bovine brain tubulin (Cyto- unbinned images (5.1 Å/pixel) to resolve the lipid bilayer. Tomograms were skeleton, Inc.), in 5 mM MgCl , 1 mM EGTA, 80 mM Pipes, 1 mM GTP at pH 2 denoised using nonlinear anisotropic diffusion (53) as implemented in IMOD 6.8, and with a tubulin monomer concentration of 50 μM, were provided by (52). Images showing the interaction between fragmented fibrils and the C. Moores (Birkbeck College, London, United Kingdom). lipid bilayer were obtained by averaging 10–30 slices of binned or unbinned, filtered tomograms, respectively. Liposome Preparation. Liposome samples of LUVs of egg PC/PG were prepared fi as previously described (27, 50). Briefly, a stock solution of 62.5 mM lipids Four representative, binned, ltered tomograms collected from three – μ prepared from PC (Type XVI-E, chicken egg, Sigma Aldrich) and PG (chicken different grids at underfocus 2.5 5 m were chosen for manual counting of fi egg, Avanti Lipids) in chloroform to give a 4:1 molar ratio of PC–PG, and contacts between bril ends or sides with liposome surfaces. Contacts were liposomes were formed by extrusion as detailed in SI Methods. GVs used in divided into three categories: end interactions, side and end interactions, confocal microscopy experiments were prepared by a rapid evaporation and side interactions only. method (51), as described in SI Methods. NBD-PE [1,2-dipalmitoyl-sn-glycero- β fi 3-phosphoethanolamine-N-(7-nitro-2–1,3-benzoxadiazol-4-yl), ammonium salt, Fluorescence Microscopy. TMR-labeled short 2m brils were 10-fold diluted − Avanti Lipids] was added to the egg PC–PG 4:1 lipid mixture at 0.04% (molar into egg PC/PG/NBD-PE (4:1:2 × 10 3, molar ratio) giant vesicle suspension, ratio) to make the vesicles fluorescent. Similar results were obtained with GVs yielding a 12 μM β2m monomer equivalent concentration and 1.8 mM total prepared with a PC–PG ratio of 1:1. lipid concentration. The images were obtained following 15 min incubation of the fibrils with the vesicles on a Zeiss Axiovert 100M confocal laser × Sample Preparation for EM. Samples of fragmented β2m fibrils and liposomes scanning microscope using a Zeiss 63 /1.4 N.A. Plan Apochromat DIC oil were prepared by adding fragmented fibrils (60 μM) (all fibril concentrations immersion objective lens. The NBD-PE fluorescent probe was excited with

are given in β2m monomer equivalent concentration) in 10 mM NaH2PO4,50 the 488 nm line of an argon laser, whereas TMR was excited with argon- mM NaCl (buffer pH 2.0) to stock solutions of PC–PG liposomes (1.3 or 2.6 krypton laser at 568 nm. Long pass filters 505 and 580 were used for ac- mM lipids in 50 mM Hepes, 107 mM NaCl, 1 mM EDTA, pH 7.4), with a final quisition NBD and TMR fluorescence, respectively.

concentration of 500–550 μM lipids and 13–18 μM β2m ﬁbrils (30–40 molar excess of lipids). A typical cryo-EM sample contained a ﬁnal concentration of ACKNOWLEDGMENTS. We thank Luchun Wang and Daniel Clare for help 0.5 mM liposomes in 50 mM Hepes, 107 mM NaCl, 1 mM EDTA, and pH 7.4, with electron microscopy and data analysis, Salvador Tomas for support in

which was vortexed for few seconds, mixed with 60 μM β2m fibril fragments sample preparation and helpful discussions, and David Houldershaw and Richard Westlake for computing support, the Hewitt and Radford group in 10 mM NaH2PO4, 50 mM NaCl, and pH 2.0 to give a final concentration of 18 μM β m fibrils. The mixture was vortexed for a few seconds and then members for helpful discussions and Giulia Zanetti for comments on the 2 manuscript. This work was supported by Wellcome Trust Grants 08059 (to incubated for 1–2 min at 25 °C. Three 5 μL aliquots of this solution were – H.R.S. and S.E.R.), 075675 (to S.E.R.), and by a Wellcome Trust equipment applied to negatively glow-discharged holey carbon grids (300 400 mesh Cu, Grant 079605 (to H.R.S.). Funding was also provided by the UK Biotechnol- fi Agar-Scienti c), manually blotted, and plunge frozen in liquid ethane. ogy and Biological Sciences Research Council (BB/526502/1) (to S.E.R.) and For cryotomography, an aliquot [12% (vol/vol) final concentration] of 10 the British Council (British Israel Research and Academic Exchange Award) nm gold beads (Protein A-gold conjugates, Aurion) was added to the sample (to S.E.R. and R.J.).

Milanesi et al. PNAS Early Edition | 5of6 Downloaded by guest on September 29, 2021 1. Hartl FU, Bracher A, Hayer-Hartl M (2011) Molecular chaperones in protein folding 27. Xue WF, et al. (2009) Fibril fragmentation enhances amyloid cytotoxicity. J Biol Chem and proteostasis. Nature 475(7356):324–332. 284(49):34272–34282.

2. Pepys MB (2006) Amyloidosis. Annu Rev Med 57:223–241. 28. White HE, et al. (2009) Globular tetramers of β(2)-microglobulin assemble into elab- 3. Wadsworth JDF, Collinge J (2011) Molecular pathology of human prion disease. Acta orate amyloid ﬁbrils. J Mol Biol 389(1):48–57.

Neuropathol 121(1):69–77. 29. Ladner CL, et al. (2010) Stacked sets of parallel, in-register β-strands of β2-micro- 4. Powers ET, Morimoto RI, Dillin A, Kelly JW, Balch WE (2009) Biological and chemical globulin in amyloid fibrils revealed by site-directed spin labeling and chemical la- approaches to diseases of proteostasis deficiency. Annu Rev Biochem 78:959–991. beling. J Biol Chem 285(22):17137–17147. 5. Buxbaum JN, Linke RP (2012) A molecular history of the amyloidoses. J Mol Biol 30. Debelouchina GT, Platt GW, Bayro MJ, Radford SE (2010) Griffin RG Intermolecular 421(2-3):142–159, 10.1016/jmb.2012.01.024. alignment in beta(2)-microglobulin amyloid fibrils. J Am Chem Soc 132:17077–17079.

6. Eichner T, Radford SE (2011) A diversity of assembly mechanisms of a generic amyloid 31. Myers SL, et al. (2006) A systematic study of the effect of physiological factors on β2- fold. Mol Cell 43(1):8–18. microglobulin amyloid formation at neutral pH. Biochemistry 45(7):2311–2321. 7. Colletier JP, et al. (2011) Molecular basis for amyloid-beta polymorphism. Proc Natl 32. Tilley SJ, Saibil HR (2006) The mechanism of pore formation by bacterial toxins. Curr Acad Sci USA 108(41):16938–16943. Opin Struct Biol 16(2):230–236. 8. Lashuel HA, Lansbury PT, Jr. (2006) Are amyloid diseases caused by protein aggregates 33. Jahn TR, Tennent GA, Radford SE (2008) A common β-sheet architecture underlies in

that mimic bacterial pore-forming toxins? Q Rev Biophys 39(2):167–201. vitro and in vivo β2-microglobulin amyloid fibrils. J Biol Chem 283(25):17279–17286. 9. Glabe CG (2006) Common mechanisms of amyloid oligomer pathogenesis in de- 34. Hebda JA, Miranker AD (2009) The interplay of catalysis and toxicity by amyloid generative disease. Neurobiol Aging 27(4):570–575. intermediates on lipid bilayers: insights from type II diabetes. Annu Rev Biophys 10. Demuro A, Smith M, Parker I (2011) Single-channel Ca(2+) imaging implicates Aβ1-42 38:125–152. amyloid pores in Alzheimer’s disease pathology. J Cell Biol 195(3):515–524. 35. Bucciantini M, et al. (2002) Inherent toxicity of aggregates implies a common 11. Butterfield SM, Lashuel HA (2010) Amyloidogenic protein-membrane interactions: mechanism for protein misfolding diseases. Nature 416(6880):507–511. mechanistic insight from model systems. Angew Chem Int Ed Engl 49(33):5628–5654. 36. Martins IC, et al. (2008) Lipids revert inert Abeta amyloid fibrils to neurotoxic pro- 12. Reynolds NP, et al. (2011) Mechanism of membrane interaction and disruption by tofibrils that affect learning in mice. EMBO J 27(1):224–233. α-synuclein. J Am Chem Soc 133(48):19366–19375. 37. Sasahara K, Hall D, Hamada D (2010) Effect of lipid type on the binding of lipid 13. Stefani M (2010) Biochemical and biophysical features of both oligomer/fibril and cell vesicles to islet amyloid polypeptide amyloid fibrils. Biochemistry 49(14):3040–3048. membrane in amyloid cytotoxicity. FEBS J 277(22):4602–4613. 38. Yoshiike Y, Akagi T, Takashima A (2007) Surface structure of amyloid-beta fibrils 14. Domanov YA, Kinnunen PKJ (2008) Islet amyloid polypeptide forms rigid lipid-protein contributes to cytotoxicity. Biochemistry 46(34):9805–9812. amyloid fibrils on supported phospholipid bilayers. J Mol Biol 376(1):42–54. 39. Knight JD, Miranker AD (2004) Phospholipid catalysis of diabetic amyloid assembly. J 15. Engel MFM, et al. (2008) Membrane damage by human islet amyloid polypeptide Mol Biol 341(5):1175–1187. through fibril growth at the membrane. Proc Natl Acad Sci USA 105(16):6033–6038. 40. Boucrot E, et al. (2012) Membrane fission is promoted by insertion of amphipathic 16. Bucciantini M, et al. (2012) Toxic effects of amyloid fibrils on cell membranes: The helices and is restricted by crescent BAR domains. Cell 149(1):124–136. importance of ganglioside GM1. FASEB J 26(2):818–831. 41. Campioni S, et al. (2010) A causative link between the structure of aberrant protein 17. Novitskaya V, Bocharova OV, Bronstein I, Baskakov IV (2006) Amyloid fibrils of oligomers and their toxicity. Nat Chem Biol 6(2):140–147. mammalian prion protein are highly toxic to cultured cells and primary neurons. J Biol 42. Winner B, et al. (2011) In vivo demonstration that alpha-synuclein oligomers are toxic. Chem 281(19):13828–13836. Proc Natl Acad Sci USA 108(10):4194–4199. 18. Mossuto MF, et al. (2010) The non-core regions of human lysozyme amyloid fibrils 43. Cremades N, et al. (2012) Direct observation of the interconversion of normal and influence cytotoxicity. J Mol Biol 402(5):783–796. toxic forms of α-synuclein. Cell 149(5):1048–1059. 19. Pieri L, Madiona K, Bousset L, Melki R (2012) Fibrillar α-synuclein and huntingtin exon 44. McMahon HT, Gallop JL (2005) Membrane curvature and mechanisms of dynamic cell 1 assemblies are toxic to the cells. Biophys J 102(12):2894–2905. membrane remodelling. Nature 438(7068):590–596. 20. Tomic JL, Pensalfini A, Head E, Glabe CG (2009) Soluble fibrillar oligomer levels are 45. Lee K-K (2010) Architecture of a nascent viral fusion pore. EMBO J 29(7):1299–1311. elevated in Alzheimer’s disease brain and correlate with cognitive dysfunction. 46. Fourniol F, Moores CA (2010) Kinesin motor activation: Microtubules pull the Neurobiol Dis 35(3):352–358. switches. Proc Natl Acad Sci USA 107(9):3949–3950. 21. Okada T, Ikeda K, Wakabayashi M, Ogawa M, Matsuzaki K (2008) Formation of toxic 47. Otomo T, et al. (2005) Structural basis of actin filament nucleation and processive Abeta(1-40) fibrils on GM1 ganglioside-containing membranes mimicking lipid rafts: capping by a formin homology 2 domain. Nature 433(7025):488–494. Polymorphisms in Abeta(1-40) fibrils. J Mol Biol 382(4):1066–1074. 48. Gosal WS, et al. (2005) Competing pathways determine fibril morphology in the self-

22. Meyer-Luehmann M, et al. (2008) Rapid appearance and local toxicity of amyloid-beta assembly of β2-microglobulin into amyloid. J Mol Biol 351(4):850–864. plaques in a mouse model of Alzheimer’s disease. Nature 451(7179):720–724. 49. Porter MY, Routledge KE, Radford SE, Hewitt EW (2011) Characterization of the re-

23. Tyedmers J, et al. (2010) Prion induction involves an ancient system for the seques- sponse of primary cells relevant to dialysis-related amyloidosis to β2-microglobulin tration of aggregated proteins and heritable changes in prion fragmentation. Proc monomer and fibrils. PLoS ONE 6(11):e27353. Natl Acad Sci USA 107(19):8633–8638. 50. Tomas S, Milanesi L (2010) Mutual modulation between membrane-embedded re- 24. Carulla N, et al. (2005) Molecular recycling within amyloid fibrils. Nature 436(7050): ceptor clustering and ligand binding in lipid membranes. Nat Chem 2(12):1077–1083. 554–558. 51. Moscho A, Orwar O, Chiu DT, Modi BP, Zare RN (1996) Rapid preparation of giant 25. Knowles TPJ, et al. (2009) An analytical solution to the kinetics of breakable filament unilamellar vesicles. Proc Natl Acad Sci USA 93(21):11443–11447. assembly. Science 326(5959):1533–1537. 52. Kremer JR, Mastronarde DN, McIntosh JR (1996) Computer visualization of three- 26. Xue WF, Homans SW, Radford SE (2008) Systematic analysis of nucleation-dependent dimensional image data using IMOD. J Struct Biol 116(1):71–76. polymerization reveals new insights into the mechanism of amyloid self-assembly. 53. Frangakis AS, Hegerl R (2001) Noise reduction in electron tomographic re- Proc Natl Acad Sci USA 105(26):8926–8931. constructions using nonlinear anisotropic diffusion. J Struct Biol 135(3):239–250.

6of6 | www.pnas.org/cgi/doi/10.1073/pnas.1206325109 Milanesi et al. Downloaded by guest on September 29, 2021